We propose to develop novel imaging techniques to quantity breast density and breast parenchymal patterns that can be used to further study the linkage of tissue composition and cancer risk. Breast density, as currently defined and measured, is a strong risk factor for breast cancer. The clinical measure of mammographic density is inadequate to study how the composition of breast tissue relates to cancer risk, especially on tissue specimens. We will develop the following: a Dual X-ray Absorptiometry (DXA) technique to quantify breast tissue composition on a clinical digital mammography device and a novel technique to quantify the projected parenchyma pattern using a combination of connectivity (Euler number) and fractal dimensions. These measures will be quantified in whole breast specimens and compared to current clinical measures of breast density (i.e., mammographic density and BI-RADS scores). In addition, we will extend these techniques to their use in thin (5 mm or less) breast tissue sections. Local measures of density and structure will be compared to whole breast measures by sectioning breasts specimens in their entirety and summing the local measures. Lastly, we will study if metabolic profiles measured using magnetic resonance magic angle spectroscopy (MR-MAS) correspond to specific pathological tissue types associated with dense breast (from Project 2) by histology analysis of the same samples. The results of this study may lead to (1) the implementation of new clinical measures of breast composition and risk in clinical mammograms, (2) novel compositional measures of density for thin tissue samples such as biopsy tissue, and (3) the identification of a metabolic signature that is specific to breast tissue type and applicable to new MR spectroscopic imaging techniques for a more sensitive and specific measure of risk.